In this week's lessons from the AI + Innovation Stage, host, Livia Bernardini, writes about taste, judgement, a solar company in Minnesota that lost a quarter of a million dollars to a machine's fantasy, and why she is chickening out. 

The AI + Innovation Stage at this summer's MAD//Fest hosted such a remarkable set of speakers. One in particular made me realise that it takes more courage than I thought to ask an LLM what it thinks of you. But let’ s start from the beginning. 

In the last couple of pieces, we covered what to focus on in the age of AI. This is key at a time when everything is telling you what to think, judgement is under pressure and decisions feel forcefully accelerated.

Today we are looking at how to do it, part one. 

I opened the stage with my keynote The Intuitive Advantage, where I urged you to train your intuition so that it can guide you alongside other data to master AI, or at least a crucial element of it, which is your judgement, so it can work better for you. I couldn't have been luckier to have Mike Pearson from Wayfair following me with a talk that seemed designed to come after mine.

Mike nailed what the industry should focus on: better briefs, better judgement, better taste and better people.

He elegantly framed the need for better taste. Taste is how you choose to do something once you have worked out what to do. I had a long play with the definition of taste and why it was so subjective to the individual, reaching the conclusion that a person's unique taste is made by what they have lived, what they see now, and what they want next. I wrote about it in "Your Best Career Hack Is The Collective".

Mike put it far more simply than I did, as Leonardo da Vinci said: "Simplicity is the ultimate sophistication." I could not have agreed more with both Mike and Leonardo.

Another set of brilliant speakers came from The AA, MSQ and The Gate team, as what they presented made me rush home to ask my husband whether we were covered by The AA. Luckily we were, and if their assistance is half as good as their new campaign and the way the team deployed it, then we are really covered and I can safely say... It's OK, I am with The AA.

They were clear that the creative idea comes first and AI works hard to protect it. Judgement, taste and the intuition required to know when something is right are what stop good work becoming slop. They had me in giggles and slightly FOMOing about their creative production tools and processes, and about the people behind it all, and about not having worked with any of them…yet.

Finally, on the how to do it part one, Dr Cecilia Dones and Iesha White told us a couple of stories that made us all sit up. The first involves a hot dog eating contest. An American professor went on Reddit and a few blogs just saying that he had won one, and within about a day Google's AI Overview was reporting the victory of this esteemed professor as an established fact. He was amused, and equally alarmed at how easy it had been.

The second story was Wolf River Electric, a solar installer in Minnesota who started losing contracts, and were told by their prospects that people did not agree with what the company was doing. They had no clue what they were supposed to have done. An AI Overview had invented malpractice suits and a handful of court cases that never happened, and by the time anyone at Wolf River worked that out, a quarter of a million dollars or thereabouts in contracts had gone down the drain.

So Cecilia and Iesha both told us to go and ask the LLMs what they think of us, and to keep asking, because the answer moves. 

Usually brave, on this one I am a chicken and I still haven't found the courage to do it. I would massively appreciate any of you who give it a go and let me know... gently please, if I should worry about anything.

If you want to say hi, connect with me on LinkedIn or follow me on Substak where I publish weekly. It’s free and I’d love to see you there!